July 29, 1875] 



NATURE 



259 



illey but its branches, would, according to Mr. Ellett, 

 • probably form a lake covering from 16 to 18 square 

 miles, with an average depth of nearly 30 feet, and con- 

 taining more than 12,000,000,000 cubic feet of water." 

 " It follows then that we should need but four dams, such 

 as we have described, to secure the valley of the Upper 

 Ohio against all destructive floods. 



This however assumes that at the beginning of a flood 

 the reservoirs will be empty— a condition on which it 

 would not be safe to rely. It also seems that the shape 

 of the valleys of the tributaries of the Ohio is everything 

 that could be wished by an engineer who desired to 

 convert them by means of dams into artificial lakes. 

 They are trough-shaped, moderately wide, long, and not 

 too steep. This last is a great advantage, because the 

 steeper the valley the shorter is the lake that will be 

 formed by a dam across it. It is likely that the Garonne 

 and its tributaries are less favourably circumstanced, but 

 nevertheless in a country of such varied contour as the 

 south-west of France, there must be many eligible sites 

 for reservoirs. In another way also the Garonne will 

 certainly be found a less manageable river than the Ohio, 

 namely that the volume of its floods bears a much higher 

 ratio to its ordinary flow. 



After the disastrous floods of the Loire in 1855, the late 

 Emperor wrote a letter to his Minister of Public Works 

 recommending the control of the floods by means of a 

 number of small reservoirs to be formed by building dams 

 across the mountain valleys. This however was lost sight 

 of, and we see the result in the ruins of Toulouse. 



A most useful work of this kind has been in operation 

 for many years in Ireland. The following particulars are 

 taken from a paper " on the Industrial Uses of the Upper 

 Bann River," by John Smyth, jun., C.E., read at the 

 Belfast Meeting of the British Association last year, and 

 ordered by the General Committee to be printed in 

 extenso. 



The purpose of the reservoirs on the Bann is not to 

 prevent floods, which, so far as we are aware, were never 

 particularly disastrous on that river, but to equalise the flow 

 of the river for water-power. "In 1835 the principal mill- 

 owners formed themselves into a provisional committee to 

 take steps to procure a better and more regular supply of 

 vvater by the construction of reservoirs. They placed the 

 matter in the hands of Sir "William Fairbairn, who, 

 assisted by J. F. Bateman, Esq., surveyed the collecting 

 grounds of the river Bann and its several tributaries." 

 Under their advice two reservoirs were constructed at 

 Lough Island Reavy and Corbet Lough. 



The Lough Island Reavy reservoir is 250 acres in 

 extent, and contains 270,000,000 cubic feet. It cost 

 15,000/. to construct, besides 6,000/. for land. It is 430 

 feet above the sea-level, and is supplied by two mountain- 

 streams. Its drainage area, including the lake itself, is 

 only about five square miles, and it is filled and emptied 

 only once in the year. 



The Corbet reservoir is lower down than the other, and 

 is chiefly filled from the Bann itself. Its extent is 70 

 acres, and its capacity 28,000,000 cubic feet. It "has 

 been of much more service than its capacity would lead 

 one to expect, as it may be filled and emptied four or five 

 times in each year by small floods in the river, and all 

 the Sunday water can be sent into it, and let down to 

 the mills on Monday and Tuesday. It is generally 

 exhausted before the upper reservoir is called on, and 

 keeps up a supply when there is a scarcity in frosty 

 weather." 



The purpose of regulating the supply has been tolerably 

 well attained. "A register of the daily height of the 

 water in Lough Island Reavy has been kept since 1847. 

 It shows that this reservoir has been of great service, as 

 during 26 years an average supplementary supply of about 

 two-fifths of the standard summer discharge allowed over 

 Ervin's Weir, or about 30 cubic feet per second, has been 



granted, for, on an average, 102 days yearly : and the 

 reservoir has been empty, on an average, eleven and a 

 half days yearly." " The register of the Corbet reservoir 

 has not been kept so long or so accurately as that of 

 Lough Island Reavy ; from the average of three years, 

 however, and comparison with the register of Lough 

 Island Reavy, Ifcalculate it has given 120,000,000 cubic 

 feet in the year, exactly one half that of Lough Island 

 Reavy, or a good supply for fifty-one days ; add to this 

 the Lough Island Reavy supply, and there is a total of 

 153 days of twenty-four hours each." "As the supply 

 from the reservoirs has only failed, on an average, eleven 

 and a half days yearly, the standard water power may be 

 said to have been almost constantly maintained : — 

 indeed it is almost as good as steam-power, but at much 

 less cost." 



The income of the Company which has made the 

 reservoirs is derived from a charge authorised by their 

 Act of Parliament of 10/. per annum per foot of fall 

 occupied by manufactories, and half of this when occu- 

 pied by flax scutching mills and country corn-mills. The 

 total fall from the upper reservoir to the bottom of the 

 lowest fall is 350 feet, of which 180 are occupied by 

 machinery. The capital of the Company is 31,000/., and 

 the dividend about three per cent., with a certainty of 

 increase, if the advance in the price of coals, and the 

 expected opening of the higher part of the district by 

 railway, lead to more of the falls being occupied. 



We think the calculations we have quoted from the 

 American engineer, and the example of what has been 

 done on a comparatively small scale in Ireland, are 

 enough to show that the most difficult problems of the 

 regulation of the flow of rivers may be approached with 

 great hope of success. 



THE GIGANTIC LAND TORTOISES OF THE 

 MASCARENE AND GALAPAGOS ISLANDS* 



II. 

 A LTHOUGH the island of Aldabra is a British pos- 

 -^"^ session, its distance from the Mauritius and the 

 Seychelles renders a supervision on the part of the Govern- 

 ment very difficult, and no control whatever can be exer- 

 cised on crews of ships who land there chiefly for the 

 purpose of cutting wood, which they require for curing 

 fish, &c. Information having reached England in the 

 course of last year that it was intended to lodge perma- 

 nently wood-cutting parties on the island, the speedy 

 extinction of the tortoises seemed imminent ; and the 

 time to prevent this seemed all the more opportune, as 

 the then Governor of the Mauritius, Sir Arthur Gordon, 

 was known to take great interest in all matters relating 

 to natural history'. Consequently the following memorial 

 was addressed to him, signed by the presidents of the 

 Royal and Royal Geographical Societies, and other men 

 of science who had made researches into the extinct fauna 

 of these islands : — 



To His Excellency the Hon. Sir Arthur Hamilton Gordon, 

 K. C. M. G.y Governor and Commander-in- Chief of Mauritius 

 and its dependencies. 



We, the undersigned, respectfully beg to call the attention of 

 the Colonial Government of Mauritius to the imminent exter- 

 mination of the Gigantic Land Tortoises of the Mascarenes, 

 commonly called " Indian Tortoises." 



2. These animals were formerly abundant in the Mauritius, 

 Reunion, Rodriguez, and perhaps other islands 6f the western 

 part of the Indian Ocean. Being highly esteeiiied as food, easy 

 of capture and transport, they formed tor .many years a staple 

 supply to ships touching at those islands for refreshment. 



* The nib>t.nnce of this article is contained in a paper read by the 

 author before tlie Rcyal Society in June, 1874, which will appear in the 

 forthcomioR volume of the " Philosophical Transactions," and to which I 

 must refer for the scientific portion and other details. Sonic facts which 

 have come to my knowledge subsequently to the reading of this paper, 

 are added. Continued from p. 339. 



